US12183452B2ActiveUtilityA1

Algorithm orchestration of workflows to facilitate healthcare imaging diagnostics

Assignee: GE PREC HEALTHCARE LLCPriority: Nov 25, 2019Filed: Nov 24, 2020Granted: Dec 31, 2024
Est. expiryNov 25, 2039(~13.3 yrs left)· nominal 20-yr term from priority
G06F 11/3698H04L 63/102G06Q 10/067G06F 9/541G16H 30/40G16H 50/20G16H 30/20G16H 15/00G06F 11/3476G16H 40/20G06F 11/0766G06F 11/079G06F 11/0751G06F 11/3664
33
PatentIndex Score
0
Cited by
23
References
24
Claims

Abstract

Techniques for orchestrating execution of algorithms on medical images according to pre-defined workflows and techniques for managing workflow and model execution are provided. In an embodiment, a system comprises a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components comprise an algorithm catalog that comprises algorithm information identifying algorithms available for processing medical images, the algorithm information comprising algorithm execution instructions for executing the algorithms as web-services; and an onboarding component that adds the algorithm information to the algorithm catalog in response to reception of the algorithm information via an onboarding user interface of an algorithm management application, wherein based on inclusion of the algorithm information in the algorithm catalog, the algorithms are made available for incorporating into workflows for executing the algorithms on the medical images.

Claims

exact text as granted — not AI-modified
What is claimed is: 
     
       1. A system, comprising:
 a memory that stores computer-executable components; and 
 a processor that executes the computer-executable components stored in the memory, wherein the computer executable components comprise:
 an algorithm catalog that comprises algorithm information identifying algorithms available for application to medical images, the algorithm information comprising algorithm execution instructions for executing the algorithms in association with invoking the algorithms at respective network accessible systems where the algorithms are stored; and 
 a workflow creation component that provides a workflow creation application that facilitates creating a workflow made available by the system for application to one or more of the medical images, the workflow comprising workflow information defining a set of computer-executable processing steps to be applied to the one or more of the medical images, 
 wherein the workflow creation application comprises a graphical user interface that facilitates creating the workflow using graphical nodes in association with receiving user input dragging and dropping the graphical nodes onto a workflow creation area of the graphical user interface, connecting the graphical nodes using one or more connection lines to define relationships between the graphical nodes, and configuring the graphical nodes, and 
 wherein the graphical nodes comprise different types of graphical nodes respectively corresponding to different types of computer-executable processing steps available for processing the medical images, including a task node corresponding to an algorithm execution task, wherein based on selection of the task node, the workflow creation application provides a selectable list of the algorithms included in the algorithm catalog and integrates a selected algorithm from the selectable list into the workflow in association with importing the algorithm execution instructions corresponding to the selected algorithm into the workflow information from the algorithm catalog, and 
 wherein the different types of graphical nodes are selected from the group consisting of: the task node, a decision node, a sub-workflow node, a wait node, a fork node, a join node and a function node. 
 
 
     
     
       2. The system of  claim 1 , wherein the computer-executable components further comprise:
 a workflow execution component that controls execution of the workflow on the one or more medical images in association with reception of an image processing request from a medical image provider via a network and returns workflow results to the medical imaging provider generated in association with the execution of the workflow, the workflow results comprising output data of the selected algorithm. 
 
     
     
       3. The system of  claim 2 , wherein at least some of the algorithms employ a different format that is not compatible with the workflow execution component, and wherein the computer-executable components further comprise:
 an onboarding component that adds the algorithm information to the algorithm catalog in response to reception of the algorithm information via an onboarding user interface of an algorithm management application, wherein based on inclusion of the algorithm information in the algorithm catalog, the system makes the algorithms available for incorporating into workflows created using the workflow creation application; and 
 a translation component that generates mapping information defining a mapping operation that controls translating between the different format and a format that is compatible with the workflow execution component in association with the onboarding, and wherein the onboarding component includes the mapping information with corresponding algorithm execution instructions for the at least some of the algorithms in the algorithm catalog. 
 
     
     
       4. The system of  claim 3 , wherein the workflow execution component employs the mapping information in association with controlling the application of the at least some of the algorithms. 
     
     
       5. The system of  claim 1 , wherein the algorithms comprise medical image inferencing models. 
     
     
       6. The system of  claim 3 , wherein the algorithms comprise medical image inferencing models, wherein the algorithm management application further provides for adjusting one or more parameters and thresholds of the algorithms in association with receiving the algorithm information, wherein the onboarding component stores settings information with the algorithm information defining the one or more parameters and thresholds, wherein the workflow execution component applies the settings information in association the execution of the algorithms, and wherein the thresholds include a critically threshold for at least some of the medical image inferencing models. 
     
     
       7. The system of  claim 1 , wherein the computer-executable components further comprise:
 a simulator component that runs the workflow in a test mode on one or more test images and identifies execution errors. 
 
     
     
       8. The system of  claim 1 , wherein the computer-executable components further comprise:
 a workflow management component that provides, via an algorithm management application, access to collective workflow information defining workflows available by the system for application to the medical images, as created via the workflow creation application, and workflow management functions related to the workflows. 
 
     
     
       9. The system of  claim 8 , wherein the workflow management functions include an activation/deactivation function that provides for selectively activating and deactivating one or more of the workflows. 
     
     
       10. A method, comprising:
 providing, by a system operatively coupled to a processor, an algorithm catalog that comprises algorithm information identifying algorithms available for application to medical images and algorithm execution instructions for executing the algorithms in association with invoking the algorithms at respective network accessible systems where the algorithms are stored; and 
 providing, by the system, a workflow creation application that facilitates creating workflows made available by the system for application to the medical images, the workflows respectively comprising workflow information defining a set of computer-executable processing steps to be applied to one or more of the medical images,
 wherein the workflow creation application comprises a graphical user interface that facilitates creating the workflows using graphical nodes in association with receiving user input dragging and dropping the graphical nodes onto a workflow creation area of the graphical user interface, connecting the graphical nodes using one or more connection lines to define relationships between the graphical nodes, and configuring the graphical nodes, and 
 wherein the graphical nodes comprise different types of graphical nodes respectively corresponding to different types of computer-executable processing steps available for processing the medical images, including a task node corresponding to an algorithm execution task, wherein based on selection of the task node, the workflow creation application provides a selectable list of the algorithms included in the algorithm catalog and integrates a selected algorithm from the selectable list into the workflows in association with importing the algorithm execution instructions corresponding to the selected algorithm into the workflow information from the algorithm catalog, and 
 wherein the different types of graphical nodes are selected from the group consisting of: the task node, a decision node, a sub-workflow node, a wait node, a fork node, a join node and a function node. 
 
 
     
     
       11. The method of  claim 10 , further comprising:
 storing, by the system, the workflows created using the workflow creation application; 
 receiving, by the system, image processing requests from medical imaging providers via a network; 
 controlling, by the system, execution of the workflows on the medical images in association with reception of image processing requests; 
 generating, by the system, workflow results based on the execution, the workflow results comprising outputs of respective algorithms integrated therein; and 
 returning, by the system, the workflow results to the medical imaging providers via the network. 
 
     
     
       12. The method of  claim 10 , wherein the algorithms comprise medical image inferencing models. 
     
     
       13. The method of  claim 10 , further comprising:
 running, by the system, the workflows in a test mode on one or more test images; and 
 determining, by the system, whether execution errors exist in the workflows based on the running. 
 
     
     
       14. The method of  claim 10 , further comprising:
 providing, by the system via an algorithm management application, access to collective workflow information defining the workflows and management functions related to the workflows. 
 
     
     
       15. The method of  claim 14 , wherein the wherein the workflow management functions include an activation/deactivation function that provides for selectively activating and deactivating one or more of the workflows. 
     
     
       16. A non-transitory machine-readable storage medium, comprising executable instructions that, when executed by a processor associated with an algorithm orchestration system, facilitate performance of operations, comprising:
 providing an algorithm catalog that comprises algorithm information identifying algorithms available for application to medical images and algorithm execution instructions for executing the algorithms in association with invoking the algorithms at respective network accessible systems where the algorithms are stored; 
 providing a workflow creation tool that facilitates creating workflows made available by the algorithm orchestration system for application to the medical images, the workflows respectively comprising workflow information defining a set of computer-executable processing steps to be applied to one or more of the medical images,
 wherein the workflow creation tool comprises a graphical user interface that facilitates creating the workflows using graphical nodes in association with receiving user input dragging and dropping the graphical nodes onto a workflow creation area of the graphical user interface, connecting the graphical nodes using one or more connection lines to define relationships between the graphical nodes, and configuring the graphical nodes, and
 wherein the graphical nodes comprise different types of graphical nodes respectively corresponding to different types of computer-executable processing steps available for processing the medical images, including a task node corresponding to an algorithm execution task, wherein based on selection of the task node, the workflow creation tool provides a selectable list of the algorithms included in the algorithm catalog and integrates a selected algorithm from the selectable list into the workflows in association with importing the algorithm execution instructions corresponding to the selected algorithm into the workflow information from the algorithm catalog, and 
 wherein the different types of graphical nodes are selected from the group consisting of: the task node, a decision node, a sub-workflow node, a wait node, a fork node, a join node and a function node. 
 
 
 
     
     
       17. The non-transitory machine-readable storage medium of  claim 16 , the operations further comprising:
 storing the workflows created using the workflow creation tool; 
 receiving image processing requests from medical imaging providers via a network; 
 controlling execution of the workflows on the medical images in association with reception of the image processing requests; 
 generating workflow results based on the execution, the workflow results comprising one or more of outputs of respective algorithms incorporated therein; and 
 returning the workflow results to the medical imaging providers via the network. 
 
     
     
       18. The system of  claim 2 , wherein the workflow execution component retrieves the one or more medical images from one or more source storage systems and loads the one or more medical images into a local storage system based on reception of the image processing request and a determination that the workflow is applicable to the one or more medical images, and wherein the execution component controls the execution of the workflow on the one or more medical images in association with accessing the one or more medical images from the local storage system. 
     
     
       19. The system of  claim 1 , wherein the execution instructions control the application of the algorithms to the medical images as web-services. 
     
     
       20. The system of  claim 1 , wherein the execution instructions control the application of the algorithms to the medical images as Kubernetes jobs. 
     
     
       21. The system of  claim 1 , wherein the different types of graphical nodes further include other types of nodes corresponding to other types of the different types of the computer-executable steps available for processing the medical images, including at least one of, a medical image filtering step, a data retrieval step, a data moving step, or a data deletion step. 
     
     
       22. The system of  claim 1 , wherein the different types of graphical nodes are visually distinctive from one another. 
     
     
       23. The system of  claim 1 , wherein the different types of graphical nodes comprise different shapes or symbols. 
     
     
       24. The system of  claim 1 , wherein the different types of nodes respectively correspond to different computer-executable code.

Join the waitlist — get patent alerts

Track US12183452B2 — get alerts on status changes and closely related new filings.

We store only your email — no account needed. See our privacy policy.